An Observationally Driven Multifield Approach for Probing the Circum-Galactic Medium with Convolutional Neural Networks

TL;DR

This paper introduces a novel convolutional neural network approach to map and infer physical properties of the circum-galactic medium using multiwavelength survey data, demonstrating improved accuracy with multifield datasets and highlighting challenges across different galaxy formation models.

Contribution

The paper develops a likelihood-free deep learning method using CNNs trained on multifield X-ray and 21-cm maps to infer CGM properties, a first in this context.

Findings

Multifield CNNs improve inference accuracy across all halo masses.

X-ray data alone is insufficient for halos with masses below 10^12.5 solar masses.

Cross-simulation training reveals challenges in generalizing models across different galaxy formation simulations.

Abstract

The circum-galactic medium (CGM) can feasibly be mapped by multiwavelength surveys covering broad swaths of the sky. With multiple large datasets becoming available in the near future, we develop a likelihood-free Deep Learning technique using convolutional neural networks (CNNs) to infer broad-scale physical properties of a galaxy's CGM and its halo mass for the first time. Using CAMELS (Cosmology and Astrophysics with MachinE Learning Simulations) data, including IllustrisTNG, SIMBA, and Astrid models, we train CNNs on Soft X-ray and 21-cm (HI) radio 2D maps to trace hot and cool gas, respectively, around galaxies, groups, and clusters. Our CNNs offer the unique ability to train and test on ''multifield'' datasets comprised of both HI and X-ray maps, providing complementary information about physical CGM properties and improved inferences. Applying eRASS:4 survey limits shows that X-ray is not powerful enough to infer individual halos with masses $\log(M_{\rm{halo}}/M_{\odot}) < 12.5$. The multifield improves the inference for all halo masses. Generally, the CNN trained and tested on Astrid (SIMBA) can most (least) accurately infer CGM properties. Cross-simulation analysis -- training on one galaxy formation model and testing on another -- highlights the challenges of developing CNNs trained on a single model to marginalize over astrophysical uncertainties and perform robust inferences on real data. The next crucial step in improving the resulting inferences on physical CGM properties hinges on our ability to interpret these deep-learning models.